Prostate cancer (PCa) is the most frequently diagnosed cancer in 112 countries with over 1.4 million new cases estimated in 2020, which is 14.1% of all new cancer diagnoses (1). Moreover, 370,000 men were estimated to die from prostate cancer in 2020 (6.8% of deaths caused by all cancer) (1). Early detection and advancement in treatments have improved survival for patients with PCa (1). However, these treatments can also have enduring adverse effects, such as the loss of lean mass and bone mass, fat mass gain, post-surgery incontinence, metabolic imbalance, and reduced quality of life (2–4).

Exercise or physical activity has been receiving attention in patient care in the oncology setting (5) due to the increasing body of research in the field of exercise oncology. Multiple epidemiological studies (6, 7) and clinical trials (8–10) consistently report improvements in health-related outcomes for PCa patients. In addition, preclinical murine model studies have also demonstrated a reduced PCa tumor volume and delayed tumor growth with an exercise stimulus (11, 12), and provide a strong mechanistic case for clinical trials to be tested on cancer outcomes. However, while numerous hypotheses exist, the mechanisms by which exercise influences tumor biology are not fully understood (13).

As such, multiple studies have been conducted to reveal the mechanisms of exercise-induced benefits for cancer patients. Alteration in circulating factors, epigenetic modulation, gene expression modulation, immune function improvement, and systemic inflammation reduction have been suggested as potential mechanisms for exercise-induced tumor suppression (13, 14). For instance, serum levels of myokines, skeletal muscle secreted cytokines and peptides, are known to be altered with skeletal muscle stimulation, and multiple preclinical studies have shown the tumor-suppressive role of myokines with direct application in different cancer cell lines, including PCa (13). Although the beneficial role of exercise in reduced disease progression, increased survival, and patient care is promising, the mechanisms underlying how exercise-induced physiological changes provide tumor-suppressive effects are not clearly understood.

One potential mechanism proposed is the involvement of extracellular vesicles (EVs) (15, 16), as exercise-induced skeletal muscle-derived EVs may reduce cancer cell proliferation and metastasis (17). EVs are small membrane-surrounded structures released from various cells (18) that transfer bioactive molecules (including DNA, RNA, and proteins) from donor to acceptor cells (19). Two main types of EVs are defined based on their cellular route of release, exosomes and microvesicles or microparticles (18). The term ‘exosome’ refers to vesicles of the endosomal system that are released through the fusion of the multivesicular body delimiting membrane with the plasma membrane, while ‘microvesicles’ or ‘microparticles’ refer to vesicles that directly pinch off the cell surface (18).

During exercise, the release of EVs packaging cytokines and myokines plays a crucial role in the communication between muscle and other tissues (20). For instance, the skeletal muscle-derived EVs have been shown to increase in response to exercise, and increased uptake of skeletal muscle-derived EVs in the liver has been shown in animal models (21). Thus, the potential role of skeletal muscle-derived EVs in reducing cancer proliferation and migration by transporting anti-oncogenic proteins and microRNAs (miRNAs) has been proposed (20). This review will provide the current evidence for the role of exercise in skeletal muscle-derived EVs concentration in the circulatory system, the alteration of EV contents, and the potential role of exercise-induced skeletal muscle-derived EVs content in PCa. In addition, we propose a potential mechanism whereby exercise enhances skeletal muscle-derived EV uptake and delivery in PCa.

EV cargoes are considered to be biologically influential in cancer progression, metastasis, and development of drug resistance (22–24). Furthermore, particular miRNAs in EV cargoes have been considered as potential diagnostic, prognostic, and predictive markers for PCa (24–26) ( Figure 1 ). For instance, miR-107, miR-130b, miR-141, miR-2110, miR-301a, miR-326, miR-331-3p, miR-432, miR-484, miR-574-3p, and miR-625 were shown to be substantially increased in circulating EVs from prostate cancer patients (n=78) compared to healthy individuals (n=28) (P<0.05) (27). Out of 11 miRNAs shown to increase in prostate cancer patients, miR-141 and miR-375 were significantly increased in patients with metastatic prostate cancer (n=16) compared to patients with localized prostate cancer (n=55) (P<0.05) (27), suggesting potential of circulating exosomes as prognostic markers for PCa. Furthermore, exosomal RNA analysis of plasma from 100 castration-resistant prostate cancer (CRPC) patients showed significantly shorter overall survival in patients with higher miR-375 and miR-1290 levels compared to those with lower miR-375 and miR-1290 levels (P=0.0045) (28). Although further validation will be required, this study also showed improved performance of predictive models for overall survival by incorporating miR-375 and miR-1290 levels with clinical prognostic factors (time to ADT failure and PSA level at the time of CRPC diagnosis) (28).

The study by Albino and colleagues (29) investigating the role of circulating miR-424 positive EVs in PCa patients demonstrated that metastatic PCa patients (metastatic castration-sensitive (mCSPC), n=16; metastatic castration-resistance (mCRPC), n=17) showed a higher frequency (P<0.05) of circulating miR-424 positive EVs compared to patients with primary tumors (n=25) and benign prostatic hyperplasia (BPH, n=6). Moreover, the application of EVs isolated from plasma of patients (n=17) in the in vitro environment showed increased tumor-sphere formation in the application of EVs from patients with mCSPC and mCRPC compared to primary or BPH patients (P<0.02) (29). In addition, the level of miR-424 containing EVs was positively associated with tumor growth in the 3D cell culture environment (tumor-sphere formation) (P=0.003) (29), suggesting a potential role of EV-contained miRNAs in PCa progression.

The study by Albino et al. (29) also generated a castration-resistant cell model using the LNCaP cell line by culturing in an androgen-depleted condition and showed significantly increased miR-424 in castration-resistant LNCaP-derived EVs compared to EVs derived from normal LNCaP cells (P<0.005), confirming the elevation in circulating miR-424 positive EVs in patients with advanced PCa (29). Furthermore, the application of EVs isolated from the castrate-resistant LNCaP-derived EVs to another PCa cell line, RWPE-1, showed increased tumor formation in a 3D cell culture model and cell migration compared to the application of normal LNCaP cell-derived EVs (P<0.005) (29), confirming the result by applying human plasma isolated EVs to cancer cell lines.

Preclinical studies involving PCa cell line-derived EVs and osteoblast cells also suggest a role for cancer cell-derived EVs in creating favorable niches for metastasis (30–32). In an in vitro study investigating the role of prostate cancer cell-induced EVs in osteoclastogensis and osteoblast proliferation, Inder and colleagues (30) demonstrated an increase in osteoclastogenesis of murine RAW264.7 pre-osteoclast cells (37 fold) and human primary-osteoblast proliferation (1.43 fold) with the presence of EVs isolated from the human prostate cancer cell line PC3 compared with no PC3- induced EVs (P<0.005). However, secreted soluble factors from PC3 cells were not able to increase osteoclastogensis of RAW264.7 cells, indicating requirement for PCa-induced EVs in osteoclastogenesis (30).

Probert and colleagues (31) reported that culture of osteoblast cells with PC-3 (high metastasis capacity to bone), C4-2 (moderate metastasis capacity to bone), and C4-2-4B (a bone metastatic lineage of C4-2) PCa cell line-derived EVs resulted in a significant increase of osteoblast cell viability (P=0.004, P=0.032, and P=0.001, respectively). In addition, co-culture of the osteoblast cells preincubated with PCa derived-EVs with PC-3 and C4-2 cells showed a significant increase (P<0.001) in PCa cell line viability compared to PCa cell lines co-cultured with osteoblast cells precultured with non- PCa cell lines (31). This study also showed a significant increase of PCa abundant miRNA in osteoblast and induced functional changes of osteoblast via EVs transported miRNA, suggesting a role of PCa cell line-derived EVs in generating metastatic niches (31).

Similarly, there is also preclinical evidence for the role of EVs in transferring integrins to different subsets of PCa cell lines (33, 34). The integrins are a diverse family of glycoproteins that allows cells to interact with extracellular matrix (ECM) molecules and, in cancer, overexpression of the integrins in cancer cells contributes to migration and invasion by disrupting the ECM molecules (35). Studies have showed increased integrins (such as α_Vβ₆ and α_Vβ₃) in PCa cell lines (PC-3 and RWPE) and PCa cell line-derived exosomes (33, 34). Incubation of α_Vβ₆ negative PCa cell line (DU145) or non-tumorigenic prostate epithelial cells with PCa cell line-derived exosomes results in de novo expression of these integrins in α_Vβ₆ negative PCa cell and non-tumorigenic prostate epithelial cells (33, 34). These results suggest that tumor-derived EVs can transfer surface proteins, especially integrins, and enhance migration and invasion of tumor cells.

EVs have also been reported to be involved in PCa drug resistance (36). Panagopoulos and co-workers (37) showed increased camptothecin (CPT, chemotherapy drug) resistance in CPT sensitive DU145 cells when EVs isolated from CPT resistant RC1 PCa cell line conditioned growth media was applied. However, when RC1 cells were cultured with EVs isolated from DU145, CPT-induced apoptosis was increased in the RC1 cells. In addition, when PCa cell lines DU-145 and 22Rv1 cells were cultured with the presence of EVs isolated from docetaxel resistance, PCa cell lines demonstrated docetaxel resistance (38). Furthermore, application of exosomes isolated from docetaxel responding PCa patients (n=6) and non-responding PCa patients (n=2) to the DU145 PCa cell line showed increased docetaxel-resistance in DU145 cells cultured with exosomes isolated from docetaxel non-responders, suggesting the potential role of tumor-cell derived EVs in drug resistance (38).

Multiple epidemiological studies in clinical oncology have consistently reported the positive impact of exercise in reducing PCa progression and enhancing survival. For instance, Kenfield and co-workers reported a 61% reduced risk (P=0.03) of PCa-related death (6) while Richman and co-workers reported a 57% reduction in disease progression (P=0.03) (7) in those with higher physical activity levels (≥3 hours/week) compared to those with lower physical activity levels. Furthermore, in preclinical studies, direct application of human serum obtained after exercise from healthy individuals to PCa cell line LNCaP showed a significant reduction in LNCaP cell growth (12). In addition, reduced tumor volume and delayed tumor growth was evident in a murine model injected with LNCaP cells exposed to human serum obtained after a bout of exercise compared with cells exposed to human serum acquired before exercise, suggesting a potential role for exercise in reducing tumor progression in vivo (12). Similar results were shown in the report by Hwang et al. (11), where human serum obtained after exercise from healthy older individuals (age > 60) was directly applied to PC-3 PCa cell lines. In addition, exercise increased blood delivery at the tumor site in a PCa animal model (R-3327 MatLyLu tumor cell orthotopically injected mice model) and reduced aggressiveness of PCa cells via reduced hypoxia at the tumor site (39) suggesting that exercise-induced physiological changes might have positive effects on cancer progression.

In addition, increased lean mass in PCa patients might also positively impact patient outcomes (10). In our recent systemic review and meta-analysis of the efficacy of exercise in improving supportive care outcomes in PCa patients with a range of treatments, improvements (P<0.001) in whole-body fat mass (-0.6 kg), lean mass (+0.5 kg), and appendicular lean mass (+0.4 kg) after exercise compared to usual care were noted (9). Furthermore, in our randomized controlled trial involving 57 PCa patients undergoing androgen deprivation therapy (ADT), 3 months of exercise significantly increased muscle strength, physical function, lean mass, and a number of patient-reported outcomes (8). Moreover, a recent retrospective report also showed that men with PCa exhibited increased prostate-specific antigen (PSA) progression-free survival and radiological progression-free survival in those with higher lean mass (P=0.03 and P<0.001, respectively) (10), providing a strong case for PCa patients to engage in exercise, especially of an anabolic nature to enhance or preserve lean mass.

Over the past decade, skeletal muscle has been identified as an important secretory organ producing a range of cytokines and peptides called myokines (40). In addition, in response to exercise training, skeletal muscle releases miRNAs into the circulation (41), and miRNAs, messenger RNAs (mRNAs), DNA, piwi-interacting RNAs (piRNAs), transfer RNAs (tRNAs), and myokines loaded into EVs for intercellular communication (20, 21). Moreover, clinical and preclinical studies have shown alteration of skeletal muscle-derived EVs concentration in the circulatory system and their contained protein levels are also altered by exercise (21, 42–48) ( Table 1 and Figure 2 ).

The importance of exercise oncology, the application of exercise medicine in cancer, has been well recognized in clinical oncology (5). Epidemiological (6, 7, 10) and clinical (8, 9) studies examining the effect of exercise in PCa patients have further established the role of exercise in patient care for men with PCa. Moreover, exercise-induced circulating factor alteration, epigenetic modulation, and gene expression have been suggested as potential mechanisms whereby exercise may impact disease progression in men with PCa (13, 14).

Recently, extracellular vesicles have been highlighted in cell-to-cell and cell-to-extracellular matrix communications, and in PCa tumor-derived EVs have been suggested as a therapeutic target (24–26). Tumor-derived EVs have potential in delivering oncogenic proteins, surface proteins, and miRNAs to non-tumor cells and contribute to progression of PCa by initiating growth-promoting signal cascades or creating metastatic niches among non-tumoral cells near cancerous cells (29, 31, 33, 34). As such, multiple proteins, miRNAs, and surface proteins in tumor-derived EVs have been identified as potential predictable markers for PCa progression (22, 23, 26). However, a few clinical exercise trials demonstrating elevation of skeletal muscle-induced EVs after exercise and preclinical studies demonstrating a potential tumor-suppressive effect of skeletal muscle-derived factors (such as myokines and miRNAs) suggest EVs as a potential delivery mechanism for skeletal muscle induced proteins and miRNAs in PCa (13, 21, 42–48, 58). Furthermore, the potential role of myokines in the facilitation of skeletal muscle-derived EV uptake in cancer has been recently proposed in a review article by Darkwah and co-workers (17).

It is important to note that research into skeletal muscle-derived EVs in cancer is at an early stage; more research is required to fully elucidate the role of exercise in PCa. For instance, the clinical exercise trials investigating the alteration of skeletal muscle-derived EVs after exercise are limited to healthy populations, limiting the generalizability of these outcomes to cancer patients. Moreover, as common adverse effects of ADT, a widspread treatment in men with PCa, are a significant loss of skeletal muscle mass and gain of fat mass (4), physiology of skeletal muscle biogenesis may differ from that of healthy subjects (13). The lack of preclinical studies investigating the direct effect of skeletal muscle-derived EVs on PCa cells also prevents a clear understanding of the role of exercise-induced skeletal muscle-derived EVs. Although skeletal muscle secretomes, such as myokines and miRNAs, have been shown to have potential in suppressing tumor growth (13, 49), the direct communication between skeletal muscle cells and PCa cells via EVs is yet to be reported.

Exercise-derived EVs have received increased attention as they provide an opportunity to understand the mechanistic benefit of exercise in cancer patients. Emerging evidence indicates that aerobic exercise affects circulating EV dynamics, including the size, morphology, and composition. Proteins, miRNAs, mRNAs, and DNAs packed in exercise-specific EVs may potentially play a role in preventing PCa development and disease progression. This review has summarized the preliminary evidence for an effect of exercise on circulatory levels of skeletal muscle-derived EV secretion, EV-containing protein or miRNA contents and the role of EV-containing factors in PCa progression, as well as the potential involvement of myokines in EV uptake in PCa. Given the speculative nature of the role of exercise-derived EVs to date, in the coming decade research will likely clarify the role of EVs with a focus on the dynamics of EVs in response to specific exercise modes and dosages, providing opportunities to enhance our understanding of the tailoring of exercise prescription on mediating possible cancer outcomes.

YZ and J-SK contributed equality to this work. YZ and T-ZW drafted the initial manuscript, and J-SK developed the manuscript. J-SK, RN, DG, RG, MH, and DT performed the revision. All authors contributed to the article and approved the submitted version.

J-SK is supported by National Health and Research Council Centre of Research Excellence (NHMRC-CRE) in Prostate Cancer Survivorship Scholarship. DG and RN are funded by an NHMRC-CRE in Prostate Cancer Survivorship (APP1116334).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.